How do I import and export via JSON?

Requires Basic

Quick answer

JSON works in both directions: export delivers your assets as a JSON array with all fields; import accepts a .json file with an array of asset objects (or a single object). As with CSV import, only asset_type and name are required per object. JSON is the better choice when you want to transfer nested values like tags (a list) or custom_fields (an object), which CSV can’t represent.

Prerequisites

Required role Logged-in user (typically an administrator or asset manager)

— Import and export are open to any logged-in user. Programmatically, a valid bearer token is enough.

JSON import/export is included in the Basic plan. The license’s asset limit applies to imports.

Guide

GUI (web interface)

1. Open the “Import & Export” section. Select the section for Data (route /data) in the navigation. The upload field notes “Supported formats: CSV, JSON”.

2. Upload a JSON file. Drag your .json file into the field or select it. A valid file contains an array of asset objects:

   assets-import.json

   [
     {
       "asset_type": "hardware",
       "name": "ThinkPad T14 – Vertrieb",
       "serial_number": "PF-3X9K2L",
       "manufacturer": "Lenovo",
       "status": "active",
       "tags": ["notebook", "vertrieb"]
     },
     {
       "asset_type": "network",
       "name": "Core-Switch 01",
       "manufacturer": "Cisco",
       "location": "Rechenzentrum / Rack A1"
     }
   ]

3. Start the import. Click “Start Import”. The result reports the number of imported records.

   Screenshot pending import-export-json-01 Uploading a JSON file in the import area with the format note CSV, JSON

   JSON files are imported through the same upload field as CSV.

4. Export as JSON. For the reverse direction, choose the format JSON for the export and download the file — it contains all fields, including tags and custom_fields.

API (curl)

Export as JSON (optionally filtered by type and status):

Export assets as JSON

curl -H "Authorization: Bearer inv_dein_token" \
  "https://demo.notory.io/api/v1/assets/export?fmt=json&asset_type=hardware&status=active" \
  --output assets.json

The file is an array of complete asset objects (the same format as the API response when creating one — including id, tenant_id, tags, custom_fields).

Import a JSON file (multipart/form-data, field file):

Import JSON

curl -X POST https://demo.notory.io/api/v1/assets/import \
  -H "Authorization: Bearer inv_dein_token" \
  -F "file=@assets-import.json"

Response

{
  "message": "Successfully imported 2 assets",
  "data": { "count": 2 }
}

Notory recognizes JSON by the .json extension. For other extensions, it first tries to parse JSON and then falls back to CSV.

Possible errors: 401 (token missing/invalid). If the file isn’t valid JSON and can’t be read as CSV either, the import fails; individual invalid objects are — as with CSV import — skipped.

What happens behind the scenes?

• Array or single object: The file may contain an array of objects or a single object. Anything else (e.g. a number or a string) is rejected.
• Same validation as CSV: Each object is checked individually against the asset schema; invalid objects are skipped, and the valid ones are created. The response only counts the successes.
• More data types than CSV: tags (a list), custom_fields (an object), numbers, and dates arrive typed — no detour through text like in CSV. Custom fields inside custom_fields are validated against the custom fields defined for the tenant.
• Roundtrip with caution: A JSON export can be imported again — unknown fields like id, tenant_id, or created_at are ignored and reassigned in that case. However, this creates new assets (the old ones aren’t updated), since there’s no duplicate check.
• Encoding: UTF-8; umlauts are not a problem.

Related topics

• How do I import and export data?
• Import assets from a CSV file
• Export your inventory for a backup
• Catalog — define custom fields — the basis for custom_fields